NBA Connections
Introduction
Motivation
So much of a positive career is determined by the circumstances surrounding it. Maybe you got the job because you were in the right place at the right time, or you missed the interview because of a flat tire and didn’t get the job. For players in the National Basketball Association, these circumstances are often experienced as what schools they attended and teams were they traded to. With this project we set out to discover if the circumstances around who players wore across their chests impacted overall career success. Did players that knew more players than others have more successful careers? Is connection early on in a basketball player’s journey - say high school or college - more indicative of a successful career?
Another question we set out to address is how do we measure success? We have decided to stick to the most quantitative measures of success, career statistics. Specifically, we are returning to the value over replacement player (VORP) statistic, but our visualizations allow for the user to choose the statistic they want to use to measure success by, from field goal percentage (FG%) to total number of rebounds (TRB).
Research Questions
- DOES CONNECTION EQUAL CAREER SUCCESS ? (measured in VORP)
- How connected are current NBA players?
- Is an early connection to NBA stars an indicator of career success?
- How do we measure success?
Background
Background on Topic
Data Sources
The datasets we utilized for this investigation were scraped from basketball-reference.com.
Findings
Success
We begin our search for these answers by attempting to quantify player success through various (mostly offensive) game statistics. As mentioned above, the standard that we, as your authors and data scientists, chose to measure success by is VORP due to the fact it is a calculated statistic that attempts to weigh many simple stats. This algorithm, which is included in full in the Reference section, attempts to coerce basketball success into one simple number by weighing the perceived importance of various statistics (such as points per game (PPG) and total rebounds per game (TRB)) in addition to the pure statistics of each player’s career. BPM, or Box Plus/Minus, is another statistic we decided to include as it is another option for a calculated box score based metric (). It is anothere weighted estimate of a player’s contribution to their team. Where BPM just looks at how much a player helps or hurts their team when they are on the court, VORP takes that and weighs by the number of minutes the player is on the court for, with the idea that better players are playing more minutes. We urge you to take a strong look at the VORP and BPM tabs especially if you are less familiar with basketball. You are also welcome to look into the purer career statistics we have included, nobably field goal percentage, PPG, and TRB.
## PhantomJS not found. You can install it with webshot::install_phantomjs(). If it is installed, please make sure the phantomjs executable can be found via the PATH variable.```
Networks
Conclusion
Limitations
- no data on AAU teams
- removed players without college & hs
Conclusion
Future Research
References
- http://curleylab.psych.columbia.edu/netviz/netviz1.html#/45
- https://www.basketball-reference.com/about/bpm2.html
Let’s clean up the format of that output:
| Speed | Distance |
|---|---|
| Min. : 4.0 | Min. : 2.00 |
| 1st Qu.:12.0 | 1st Qu.: 26.00 |
| Median :15.0 | Median : 36.00 |
| Mean :15.4 | Mean : 42.98 |
| 3rd Qu.:19.0 | 3rd Qu.: 56.00 |
| Max. :25.0 | Max. :120.00 |
In a study from the 1920s, fifty cars were used to see how the speed of the car and the distance taken to stop were related. Speeds ranged between 4 and 25 mph. Distances taken to stop ranged between 2 and 120 feet, with the middle 50% falling between 26 and 56 feet.
You can also embed plots as normal, for example:
Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.
